The Mysterious Third

Has it already been a month since I last discussed Adorjan? Judging from the post 'Black Is OK' - 12 Discussion Points, so it would appear. A week after that post I looked at Adorjan's first three discussion points on my chess960 blog -- Random Position, Random Results? -- where I concluded,

Getting back to traditional chess, the foundation of the 'Black Is OK' philosophy rests entirely on the second point. To quote the original statement in full: 'Qualified players will mostly come up with the same reply as a great number of world champions or chess thinkers since Lasker: the logical outcome of the game is a draw.'

What do the other nine points add to the discussion? The fourth point on which I quoted Adorjan was

Portisch said something to the effect that he had tried almost all openings and defences during his long chess career, and found that about two-thirds of these were disadvantageous for BLACK. So there is the remaining one-third, and all BLACK has to do is play these openings and defences, and then he has nothing to fear.

This is a mysterious, thought provoking statement. First, how do we count openings and defenses? ECO has 500 top level codes (5 volumes A-E with 100 codes per volume). Does this mean that only 170 of these codes are good variations for Black? No, because variations that are balanced for both sides receive far more attention than variations that are clearly much better for White.

GM Lajos Portisch, an eight-time candidate for the World Championship, contributed a chapter to Adorjan's third book, 'Black Is OK Forever!' (2005): 'Lajos Portisch: BLACK IS OK if s(he) finds the right lines!'. There he wrote,

I have tried practically all normal defences with BLACK during my long career. (Except for such defences as the Scandinavian, Alekhine, and Philidor. I 'respect' the two latter ones only because the namegivers were great figures in the history of chess [...] but I am still not willing to play either of them.) In my opinion, at least two-thirds of all 'tested' openings give White an apparent advantage. But do not ask me, dear reader, to name these systems. Considering all this, it is logical that statistics show White's advantage in the final account. (p.109)

The Scandinavian (1.e4 d5; ECO B01) and Alekhine (1.e4 Nf6; B02-05) defenses together account for five ECO codes. The Sicilian (1.e4 c5; B20-99) accounts for 80 codes. The three opening moves represent 15% of Black's 20 legal responses to 1.e4. Some of the 80 Sicilian ECO codes are unsuitable for Black (and a few are unsuitable for White). The task of counting openings and defenses seems hopeless.

Rather than identify one-third of something that can't be easily quantified, I decided to look at what Portisch really played. I happen to have a PGN collection of his games and loaded it into a chess database (SCID). It counted 2738 games played through the year 2000, and let me derive the following chart. The numbers (like 'x99') are game counts; for example, in response to 1.e4, Portisch played 1...c5 in 306 games.

From this we see that Portisch, who was known as a thorough analyst and well-prepared opening theoreticion, played the Sicilian Najdorf, the Sicilian Kan (Paulsen), and the Closed Spanish after 1.e4. He played the 1...Nf6/2...e6 complex (including the Queen's Gambit) after 1.d4.

I also thought it would be useful to look at the systems Portisch played as White. He was primarily a 1.d4/1.c4/1.Nf3 player, and avoided the 1...Nf6/2...g6 complex as Black. How he played against it as White is shown near the bottom of the diagram. In 'Presumption of Innocence', Adorjan wrote,

Lajos is a great player and he is right in most cases, the element of subjectivity is present with him as well as any of us. I saw him, for example, experiment with the Gruenfeld, and he could not make heads or tails of it. Somehow it was not his cup of tea. On the other hand, I have also seen him play strategies I considered much more difficult with a devastating effect.

The preceding analysis exposes only the tip of the iceberg. In each of Portisch's preferred opening systems -- the Najdorf, the Kan, etc. -- there are good moves to be played by Black and bad moves to be avoided. Ditto for White. No one ever said that chess was simple.